Sciencemadness Discussion Board

aluminium oxide crystal

D4RR3N - 6-9-2011 at 14:21

How can I grow an aluminium oxide crystal without the need of actually melting aluminium oxide (melts at over 2000C!)

What temperature would aluminium hydroxide decompose into aluminium oxide?

AJKOER - 6-9-2011 at 18:49

One source notes that by passing NH3 into an aqueous Al salt solution, a voluminous amorphous substance called Aluminum oxide hydrate is formed. This slowly changes via boehmite and bayerite into the thermodynamically stable hydragillite.

Gamma Al2O3 can then be prepared by the cautious dehydration of hydragillite or boehmite. Heating over 1000 C produces Alpha Al2O3, noted for its hardness.

Alpha Al(OH)3 (which can be dehydrated to Al2O3, I would guess the Alpha form) can be directly prepared by passing CO2 through Sodium aluminate solution at 80 C, upon which crystalline Alpha Al(OH)3 forms directly. If, however, the reaction is below room temperature or performed too quickly, bayerite is first formed which slowly converts into Alpha Al(OH)3.

I hope this helps.

[Edited on 7-9-2011 by AJKOER]

[Edited on 7-9-2011 by AJKOER]

Neil - 7-9-2011 at 04:07

You can grow tiny crystals by seeding a pool of aluminum hydroxide slime and waiting a long time - AJKOER seems to have that covered.

But if you are trying for sizable pieces try scrapping them off of garnet sand paper.

AFAIK there are no ways to grow saffire or ruby at home without a vacuum kiln.

blogfast25 - 7-9-2011 at 04:27

Quote: Originally posted by D4RR3N  
How can I grow an aluminium oxide crystal without the need of actually melting aluminium oxide (melts at over 2000C!)

What temperature would aluminium hydroxide decompose into aluminium oxide?


It maybe possible to dry Al(OH)3 (to Al2O3) by means of azeotropic distillation. I recall reading about that for drying Zr(OH)4 but can't recall the azeotroping solvent that was used. At most that will yield micro-crystals, not macro-crystals.

I don't think growing larger crystals of annealed alumina is possible w/o prolonged high temperatures. If it was easier we'd all be producing sapphires and rubies. Another GRQ scheme thwarted! :D

D4RR3N - 7-9-2011 at 11:45

I was looking for larger crystals, Why would you need a vaccum furnace, it is an oxide?

Neil - 7-9-2011 at 12:22

I over stated sorry, an inert gas filled furnace works but you need to flush it out very well and that much gas costs $$$

You are way above the realm of NiCrome wire for melting Alumina - you need graphite elements. Any O2 left in the kiln when it is heated will start burrowing through the elements and transporting carbon into the alumina.


Vacuum is just cheaper then massive amounts of inert gasses. What do you need these for? You can buy large pieces of factory made saph/ruby for a lot less then you can build the furnace, heat controller and what not to do it.

I tried to crack that nut, it cracked me.

IrC - 7-9-2011 at 14:24

The only method I know of starts with a large melted mass of Al2O3 (nonstop heat), a seed crystal, and a very slow moving rod which pulls the crystal out of the mass as it grows. I have seen nice looking Ruby rods for lasers made this way, by adding some Chromium doping.


Edit: forgot - in a vacuum.


[Edited on 9-7-2011 by IrC]

unionised - 8-9-2011 at 10:12

This is probably the least implausible way to grow alumina crystals a home.
http://en.wikipedia.org/wiki/Verneuil_process

AJKOER - 8-9-2011 at 13:28

Actual at home a long time ago, my friend and I played with an Atomic Hydrogen flame by simply passing generated H2 through an electrical arc.

The flame temperature actually exceeds what is required for the production of synthetic gems. The chemical issue is providing a good sustainable production source for H2 gas. My friend handled the electronics.

Neil - 8-9-2011 at 13:31

alumina is an electrical insulator...

unionised - 9-9-2011 at 09:09

Quote: Originally posted by Neil  
alumina is an electrical insulator...



So...?

Neil - 9-9-2011 at 09:27

Quote: Originally posted by AJKOER  
... played with an Atomic Hydrogen flame by simply passing generated H2 through an electrical arc.

The flame temperature actually exceeds what is required for the production of synthetic gems...


@unionised;

So you can not use a hydrogen arc torch... :o

D4RR3N - 9-9-2011 at 11:57

I have a vacuum pump and a cylinder of argon, I have a fused quartz tube, got some ceramic fibre somewhere too so half way there. Now these graphite elements look interesting but how much would the controller cost? I don’t think the elements themselves are going to be that expensive but I bet the controllers are!

Neil - 9-9-2011 at 19:13

1600-1725 °C wiki melting point range for Quartz



2072 °C - wiki melting point for Alumina


Panache - 9-9-2011 at 20:57

decent tubes of al2o3 can be got from new or used sodium vapour lamps. When i kept melting the point oxygen feed on a glassblowing torch i was playing with i replaced the stainless one with an alumina one from the lamp mentioned (the reason for my genius was that it happened to be sitting directly in front of me on the bench). It is a reasonably easy material to shape, i was using a dremel, i expected it to be far more brittle than it was. Anyway it glows bright on the torch when i use it in the way that i was, did the job.
If you know anyone who grows pot(or hydroponic tomatoes) ask them for an old lamp then jut cut off the glass outside or smash it and voila.

simba - 24-9-2011 at 10:21

Are rubies produced by the Verneuil Process as good as natural ones? Because the process seems yet to simple to me.

bobm4360 - 24-9-2011 at 19:42

Better! There aren't any inclusions. The process at lab scale is illustrated in Strong's "Procedures in Experimental Physics". This book is not about physics experiments, but about building apparatus. There are chapters on glassblowing, high vacuum, high temperatures, and others. It's dated (1938), but very readable and well illustrated.

Regards,
Bob

Paddywhacker - 24-9-2011 at 22:28

I have heard that small rubies can be produced as a side-product of the thermite reaction. You would need a thermite mix that didn't spray the alumina out of the crucible, so ... a slow burning mix.

Maybe the addition of a flux could help the molten alumina to coalesce into larger blobs. Suitable fluxes could be determined experimentally, but might include sodium chloride, sodium carbonate, ground glass, silica powder ...

unionised - 25-9-2011 at 01:43

Quote: Originally posted by Neil  
Quote: Originally posted by AJKOER  
... played with an Atomic Hydrogen flame by simply passing generated H2 through an electrical arc.

The flame temperature actually exceeds what is required for the production of synthetic gems...


@unionised;

So you can not use a hydrogen arc torch... :o


Why not?
I can do soldering (at about 200C) with a blowtorch (about 2000C)

turd - 25-9-2011 at 02:34

Quote: Originally posted by IrC  
The only method I know of starts with a large melted mass of Al2O3 (nonstop heat), a seed crystal, and a very slow moving rod which pulls the crystal out of the mass as it grows. I have seen nice looking Ruby rods for lasers made this way, by adding some Chromium doping.

Quote: Originally posted by unionised  
This is probably the least implausible way to grow alumina crystals a home.
http://en.wikipedia.org/wiki/Verneuil_process

What IrC describes sounds more like Czochralski than Verneuil. If the second post was a response to the first, that is.

CVD is performed far below the melting point of alumina. Practicable in a home setting? I doubt it.

Edit: Hydrothermal is another possibility which might be the most accessible in a home setting. Though the conditions needed according to a quick Google search are not encouraging (supercritical).

[Edited on 25-9-2011 by turd]

Neil - 25-9-2011 at 03:36

Quote: Originally posted by unionised  
Quote: Originally posted by Neil  
Quote: Originally posted by AJKOER  
... played with an Atomic Hydrogen flame by simply passing generated H2 through an electrical arc.

The flame temperature actually exceeds what is required for the production of synthetic gems...


@unionised;

So you can not use a hydrogen arc torch... :o


Why not?
I can do soldering (at about 200C) with a blowtorch (about 2000C)




...because alumina is an insulator?

How are you going to get an arc to travel preferentially to a material that is not conductive? Nothing to do with temperature.

I recently talked to a fellow who made some small rubies in a electric furnace using this method

http://sciencelinks.jp/j-east/article/200601/000020060105A09...



And from below
http://pubs.acs.org/doi/abs/10.1021/ja049678v

"the crystal growth was conducted by heating a mixture of solute (Al2O3 + 0.5 mass% Cr2O3) and flux (MoO3) at 1100 °C, followed by holding the solution at this temperature for 5 h."


1100°C it is much more in the range of DIY gear


@D4RR3N, Sounds like you might already have the stuff to do it.

Neil - 29-9-2011 at 06:01

A bee in my bonnet...

I used to have a small furnace with a roughly 5Al2O3/SiO2 refractory and Cr2O3 doping.

It had a ~2"x8" bore and mainly was run on propane, having been built to hold sawed off CO2 cartridges for micro melts. I used the same burner on it that I used on a 6"X12" furnace, to point, it was very hot.

It ended up being primary used to sinter test pieces of ceramic, but not before it was contaminated with some borax flux and wood ash.

I still have the mini plinth block turned kiln table that I used for all of the furnaces life and curiously there is a layer of clear running to pink/red glass smeared on one side. The smear is around five years old so I'm not positive I remember correctly but IIRC it was caused by borax landing on the plinth however, it may also have been wood ash.

At any rate the smear was not sticky once it had been super heated during a crucible sintering attempt so I left it and never thought about it till now.

I can not scratch it with a uncoated hacksaw blade or a bastard file. It scratched the carbide on a wood saw blade and seemed to be scratched by the carbide but only very lightly.

The only data on the blade about the carbide is that the blade was "Manufactured from quality steels and carbides" I'd guess it would be C1 or C2?

Under a microscope at 100X it has no distinct crystals but does contain lots of bubbles.

It has taken heats of over 3000°F many times and is still around, it cuts glass with ease.

I have nothing with a known hardness to test it with :(

There was obviously carbon in the furnace and the formation of true ruby is far to finicky IMHO to be formed by crude accident, which leaves me at a loss to explain it.

Thoughts? Could it be crystalline alumina?


I only have a small sample so I'm loath to start boiling it in acids, thoughts on how to non-destructively identify it?

Wizzard - 29-9-2011 at 09:44

I think I found my next project :)

MoS2 is converted to MoO3 in bake with oxygen rich environment, so I'll do that after drawing a vacuum after purging with an inert gas, and then the MoO3 is water soluble- Excellent way to get it out of the mix :) Then dry, and dessicate, recrystallize and dessicate again. Or I suppose I could just purchase it.

Then, pop it all in a quartz tube, fire it up, and hope for some results! I'll likely boil out all the MoO3 into a cooler vessel. From there, analyze results, and improve! I think my crystal lust is reaching a peak here, next I'll only want to make diamond!

peach - 29-9-2011 at 23:51

Quote: Originally posted by Neil  

...because alumina is an insulator?

How are you going to get an arc to travel preferentially to a material that is not conductive? Nothing to do with temperature.


The arc goes between two electrodes and through hydrogen, then the hydrogen transfers the heat to a none conductive surface;

Quote:
Atomic hydrogen welding (AHW) is an arc welding process that uses an arc between two metal tungsten electrodes in a shielding atmosphere of hydrogen. The process was invented by Irving Langmuir in the course of his studies of atomic hydrogen. The electric arc efficiently breaks up the hydrogen molecules, which later recombine with tremendous release of heat, reaching temperatures from 3400 to 4000 °C. Without the arc, an oxyhydrogen torch can only reach 2800 °C.[1] This is the third hottest flame after cyanogen at 4525 °C and dicyanoacetylene at 4987 °C. An acetylene torch merely reaches 3300 °C. This device may be called an atomic hydrogen torch, nascent hydrogen torch or Langmuir torch. The process was also known as arc-atom welding.

The heat produced by this torch is sufficient to melt and weld tungsten (3422 °C), the most refractory metal. The presence of hydrogen also acts as a gas shield and protects metals from contamination by carbon, nitrogen, or oxygen, which can severely damage the properties of many metals. It eliminates the need of flux for this purpose.

The arc is maintained independently of the workpiece or parts being welded. The hydrogen gas is normally diatomic (H2), but where the temperatures are over 600 °C (1100 °F) near the arc, the hydrogen breaks down into its atomic form, simultaneously absorbing a large amount of heat from the arc. When the hydrogen strikes a relatively cold surface (i.e., the weld zone), it recombines into its diatomic form and rapidly releases the stored heat.


The bigger problem there might be having an excess of atomic hydrogen present at thousands of degrees, since the goal is forming, or melting, oxides.

A Browns gas generator may be a better idea, as you'll get a balanced atmosphere out of that.

Hydrogen is not a great thing to have in a weld either, as that'll also embrittle the join. Which is part of the reason why GTAW (TIG) with Argon is used now.
Quote: Originally posted by shivas  
Are rubies produced by the Verneuil Process as good as natural ones? Because the process seems yet to simple to me.


The equipment is certainly more basic than the CVD chambers, but I have a sneaking suspicion it may be fairly difficult to get a clean lump of ruby out of it. It'll probably need at least a few goes to get it all tuned.

I expect the first few goes will result in a melted support and chamber or the ruby integrated into the support.

If you have a hydrogen cylinder or generator, give it a go. Take photos!

That'd make a wedding ring worth boasting about, home made gems and home refined precious metals.

Not so sure about this though, diamonds made from human ashes.

"Made it myself...", "At a jewellery class?", "From my dead cat..."

------------------------------------------------------------------

Irving Langmuir


Atomic hydrogen torch, this article has details of the current, voltage, electrode sizes and further design elements of the device.



Buy two of these


And a roll of that


A high tech atomic hydrogen source


You can see a Czochralski crystal furnace in operation towards the start of this video;

<iframe sandbox width="640" height="480" src="http://www.youtube-nocookie.com/embed/aWVywhzuHnQ" frameborder="0" allowfullscreen></iframe>

[Edited on 30-9-2011 by peach]

Neil - 30-9-2011 at 05:11

Thank you Peach, sorry Unionised; I was under the incorrect impression it was more like a heli-arc set up.



Browns gas may indeed be wonderful for melting alumina, its oxidizing flame may be useful in the material prep as well as the actual melting. Only most Browns gas torches I've seen people build have a high rate of self destruction... :(

watson.fawkes - 30-9-2011 at 07:26

Chapter 12 of Strong's Procedures in Experimental Physics has a lovely diagram of a Verneuil process apparatus, detailed enough to build from, though much less than a measured diagram or a kit. Unfortunately, it doesn't have any commentary on how to operate it, not at least that I could recall nor quickly locate.

simba - 30-9-2011 at 07:38

Quote: Originally posted by peach  

Not so sure about this though, diamonds made from human ashes.

"Made it myself...", "At a jewellery class?", "From my dead cat..."


I remember seeing this on tv like 10 years ago, so yes its for real. Prices are high but acceptable, at least when I saw it.

peach - 30-9-2011 at 17:49

Quote: Originally posted by Neil  
Browns gas may indeed be wonderful for melting alumina, its oxidizing flame may be useful in the material prep as well as the actual melting. Only most Browns gas torches I've seen people build have a high rate of self destruction... :(


That's what they get for messing with over-unity and defying the grid assassins.
Quote: Originally posted by shivas  

I remember seeing this on tv like 10 years ago, so yes its for real. Prices are high but acceptable, at least when I saw it.


My UK Internet humour has scuppered this situation. I know it's possible (since anything organic will turn to diamond when roasted and crushed at those values), but it's a bit of an odd thing. In some ways it's very special and personal, in others, it's weird.

[Edited on 1-10-2011 by peach]

watson.fawkes - 1-10-2011 at 05:13

Quote: Originally posted by peach  
In some ways it's very special and personal, in others, it's weird.
Not nearly as weird as it could be. They could be making graphite for pencils. "Commemorate your loved one with words: your words, their body."

Neil - 2-10-2011 at 13:19

Quote: Originally posted by peach  
Quote: Originally posted by Neil  
Browns gas may indeed be wonderful for melting alumina, its oxidizing flame may be useful in the material prep as well as the actual melting. Only most Browns gas torches I've seen people build have a high rate of self destruction... :(


That's what they get for messing with over-unity and defying the grid assassins.


:D
Quote:

The late distinguished, most excellent Professor Ebelmen made numerous experiments upon the subject of artificially-crystallized alumina. He exhibited specimens received from him years ago. They are exceedingly small, but remarkably characteristic. Here are some small scales of alumina crystallized by the operation of fire. The process consists of heating alumina, mixed with about three or four times its weight of borax, in a small platinum crucible, under certain precautions, in a porcelain furnace—a porcelain furnace where we obtain hard porcelain, and not the so-called porcelain which we make in this country, and which is made at a low temperature. It must be the true China porcelain furnace, such as was in operation at Sevres, where he experimented. He exposed the borax and alumina at a high temperature in this furnace. The borax dissolved the alumina, and after a time off went the boracic acid and the sodium. The crystallized alumina remained at last in brilliant colourless scales.


From http://books.google.ca/books?pg=PA149&lpg=PA149&dq=c...



simba - 2-10-2011 at 15:13

Quote: Originally posted by watson.fawkes  
Quote: Originally posted by peach  
In some ways it's very special and personal, in others, it's weird.
Not nearly as weird as it could be. They could be making graphite for pencils. "Commemorate your loved one with words: your words, their body."


Or charcoal also haha...imagine making a barbecue with your parent's ashes, I wonder if it could be considered cannibalism.

[Edited on 2-10-2011 by shivas]

MeSynth - 3-10-2011 at 21:45

Quote: Originally posted by D4RR3N  
How can I grow an aluminium oxide crystal without the need of actually melting aluminium oxide (melts at over 2000C!)

What temperature would aluminium hydroxide decompose into aluminium oxide?


Chances are your going to have to find something that it's soluble in. I know that a quick way to make it is by mercury aluminum amalgam but then your working with mercury. This sounds like something I would just give up on..

Morgan - 2-3-2014 at 16:25

I would like to have a large thin sheet or disk about 50 cm in diameter for some electrical experiments.
Sapphire Screen: The Making of A Scratch-Proof Smartphone
https://www.youtube.com/watch?v=vsCER0uwiWI#t=55s

bfesser - 2-3-2014 at 17:04

I'll assume you meant 50 mm and not 50 cm.

<a href="http://www.ebay.com/itm/261388183169" target="_blank">SAPPHIRE WAFER WINDOW for OPTICAL LASER OPTICS APPLICATIONS BIN#5K-10</a> <img src="../scipics/_ext.png" />

Morgan - 2-3-2014 at 18:21

No, I was dreaming. Maybe in 20 years the costs will come down.

Sapphire sheets
Class400 ha
http://www.crystals.saint-gobain.com/uploadedFiles/SG-Crysta...

Some pretty pictures if you scroll down.
http://arcturus.ligo.caltech.edu/docs/public/P/P030027-00.pd...

http://tech.fortune.cnn.com/2014/02/27/apple-sapphire-arizon...

[Edited on 3-3-2014 by Morgan]

chornedsnorkack - 3-3-2014 at 04:00

Quote: Originally posted by D4RR3N  
How can I grow an aluminium oxide crystal without the need of actually melting aluminium oxide (melts at over 2000C!)

What temperature would aluminium hydroxide decompose into aluminium oxide?


Under 220 Celsius. Seriously:
http://www.minsocam.org/ammin/AM57/AM57_1375.pdf

But just because corundum is stable over 220 Celsius, and under 220 Celsius at humidity/activity under 1 bar, does not mean diaspore would rapidly convert to corundum under these conditions. Nor is dry heat in solid conducive to rapid growth of big monocrystals.

You need a solvent. And I wonder about a good one.

Water itself is not liquid under 220 Celsius and 1 bar, and increasing pressure shifts equilibrium towards diaspore, so that corundum can precipitate from water from 360 degree temperature and 190 bar pressure.

But the problem is, the solubility of alumina in neutral water is poor, and that will hamper crystal growth.

Solvents? Alumina being amphoteric, both bases and acids are good solvents.
But the problem is that they also form salts in solid forms.

For example, you can achieve 330 Celsius in 1 bar sulphuric acid... and the water activity in 98 % sulphuric acid is much below 1 bar, so you could get out of diaspore stability field in boiling concentrated sulphuric acid. But the problem is, you are then not in stability field of corundum... you are in stability field of aluminum sulphate, or perhaps aluminum hydrogen sulphates.

If you produce a boiling, saturated solution of aluminum sulphate (note that the solution will have its boiling point raised by solute, and the heating further increases aluminum sulphate solubility, which again raises boiling point etc.), and boil it with aluminum hydroxide precipitate, will you be growing diaspore crystals or will aluminum sulphate form solid crystalline basic salts?
What is the boiling point and water activity of saturated boiling aluminum sulphate solution?

DraconicAcid - 3-3-2014 at 10:41

Quote: Originally posted by chornedsnorkack  
Quote: Originally posted by D4RR3N  
How can I grow an aluminium oxide crystal without the need of actually melting aluminium oxide (melts at over 2000C!)

What temperature would aluminium hydroxide decompose into aluminium oxide?


Under 220 Celsius. Seriously:
http://www.minsocam.org/ammin/AM57/AM57_1375.pdf

But just because corundum is stable over 220 Celsius, and under 220 Celsius at humidity/activity under 1 bar, does not mean diaspore would rapidly convert to corundum under these conditions. Nor is dry heat in solid conducive to rapid growth of big monocrystals.

You need a solvent. And I wonder about a good one.


I have spent a long time thinking about this particular project, and it seems to me that a molten salt would probably work. Aluminum hydroxide is not likely to be soluble even in a melt, but one could conceivably form crystals of alumina by dissolving some other aluminum salt in a melt, along with some anion that would slowly decompose to give oxide ion (sulphite? carbonate?).

Alternatively, one could melt something like aluminum isopropoxide, and hold it at a temperature where it slowly decomposes to eliminate the organic parts. It would have to be a slow decomposition, or you'll never get crystals.

blogfast25 - 3-3-2014 at 11:28

Quote: Originally posted by DraconicAcid  
I have spent a long time thinking about this particular project, and it seems to me that a molten salt would probably work. Aluminum hydroxide is not likely to be soluble even in a melt, but one could conceivably form crystals of alumina by dissolving some other aluminum salt in a melt, along with some anion that would slowly decompose to give oxide ion (sulphite? carbonate?).



Alumina (not the hydroxide) is of course soluble in stuff like molten Na3AlF6 and also CaF2.

Boffis - 3-3-2014 at 11:30

I think you are all missing the obvious answer. The OP wanted to know how to grow Al2O3 crystals; answer from a suitable solvent.

"Suitable solvents" for are a bit thin on the ground but it dissolves in molten cryolite Na3AlF6 and also PbF2 (Mp c 820 C) and PbO (Mp c 900 C) and there may be other, eutectic, mixture with lower melting points. The vapour pressure of molt lead fluoride is high so it evaporates fairly quickly at >1000 C and this method is also used for crystal growing. Cryolite has a higher melting point but you may be able to reduce this by adding other alkali metal fluorides like lithium.

All you need then is a high temperature furnace with a good temperature control unit that allows controlled slow cooling and one of those platinum-iridium crucibles.

DraconicAcid - 3-3-2014 at 12:03

Quote: Originally posted by blogfast25  
Quote: Originally posted by DraconicAcid  
I have spent a long time thinking about this particular project, and it seems to me that a molten salt would probably work. Aluminum hydroxide is not likely to be soluble even in a melt, but one could conceivably form crystals of alumina by dissolving some other aluminum salt in a melt, along with some anion that would slowly decompose to give oxide ion (sulphite? carbonate?).



Alumina (not the hydroxide) is of course soluble in stuff like molten Na3AlF6 and also CaF2.


True, but I was thinking of lower-temperature melts that would be easier to work with at home. I've actually worked with cryolite melts, and they weren't fun.

blogfast25 - 3-3-2014 at 12:12

Considering the Standard Enthalpy of Formation (and that most of that is actually lattice energy) finding a low temp. ionic solvent that can rip that lattice apart will always be a challenge. Never say 'never', though (ooops!)


[Edited on 3-3-2014 by blogfast25]

DraconicAcid - 3-3-2014 at 12:27

Quote: Originally posted by blogfast25  
Considering the Standard Enthalpy of Formation (and that most of that is actually lattice energy) finding a low temp. ionic solvent that can rip that lattice apart will always be a challenge. Never say 'never', though (ooops!)


Yes....which is why I was suggesting forming it from an aluminum salt and a source of oxide ion in the melt, rather than dissolving alumina in the melt.

chornedsnorkack - 4-3-2014 at 02:55

As I pointed out, aqueous solutions have the problem of boiling in diaspore stability field.
Cryolite/fluoride melts melt over 900 Celsius, which is much more than the needed temperature of 200...400 Celsius.
Common salts which do melt under 400 Celsius include alkali. But alkali would not precipitate alumina seeing that the stable solid phase would be solid aluminates.

However, salts that also are low melting include nitrates of alkali metals.
Is alumina soluble in alkali metal nitrate melts?
If aluminum nitrate were decomposed in alkali metal nitrate/nitrite melts, would it be a good condition to crystallize alumina, or would alumina form aluminates?

rskennymore - 29-4-2014 at 14:38

Has anyone had any luck attempting this?

I read an ACS publication from some folks from japan a few months back and gave it a shot with some cheaper equipment...
link: https://www.jstage.jst.go.jp/article/jcersj/113/1323/113_132...

The gist of it is they're using Molybdenum Trioxide as a flux to dissolve the alumina/chromium, which then slowly evaporates off leaving nice tetragonal crystals.

The experiment: 28.5 grams of Molybdenum Trixoide was mixed with 1.5 grams of Aluminum Oxide doped with about 0.5%wt Cr2O3. All reagents were ACS Reagent grade 99.5% or higher.

Since Platinum crucibles are out of my budget range for a random project, I decided to attempt this experiment using a cheaper zirconia crucible I found on ebay.

The crucible, with lid slightly offset to facilitate vapor escape, was placed in a small test kiln with NiChrome heating elements and heated to 300C for 2 hours and then slowly raised to 1100C over the next 3 hours. After reaching 1100C it was held there for 5 hours and then the kiln was powered off and allowed to cool overnight.

At some point through the cooling process the crucible cracked due to the large amount of residue present in the bottom with an obviously different thermal expansion ratio. The Japanese experimenters indicated that after 5 hours ~99% of the flux had evaporated so this was not anticipated and leaves me wondering what kind of ventilation their furnace had. Upon further inspection it became apparent that the zirconia had been attacked by the flux, diffused into the melt, and possibly prevented the flux from evaporating.

However, after removing the solid mass from what was left of the crucible and grinding it up in a mortar and pestle, tiny red sand-grain sized crystals were found and separated from the gravel via tweezers. They aren't big, but they scratch glass and are red, so I am calling this a success.

Pictures to prove it: http://imgur.com/a/KXLAg#0

I can say definitively that zirconia is the wrong material to use for this process. Iridium is most likely the correct material. I have attempted it since this experiment using a chemically bonded silicon carbide crucible, and it appears to also be attacked by Molybdenum Trioxide flux. I'd like to give Boron Nitride a try but haven't got the budget this month...

Cheaper materials that I think *might* work would be tantalum, niobium, and maybe a nickel-chromium crucible if you could somehow plate it with iridium or something. I aim to try the latter at some point in the near future.

Working on a source for Iridium Chloride if anyone has any leads ;)

Cheers

deltaH - 30-4-2014 at 03:33

Well done rskennymore! Your ruby grains are beautiful.

I wonder if zirconium crucibles might be an alternative. I used to use them for sodium peroxide fusions at red heat to oxidise mineral samples for metal assays. They were tough as nails.

rskennymore - 30-4-2014 at 11:32

Quote: Originally posted by deltaH  
Well done rskennymore! Your ruby grains are beautiful.

I wonder if zirconium crucibles might be an alternative. I used to use them for sodium peroxide fusions at red heat to oxidise mineral samples for metal assays. They were tough as nails.


Not a bad idea. Zirconium is pretty reactive with oxygen at higher temps from what I gather. I wonder how long it would last in atmosphere at those temps? It *shouldn't* pull oxygen out of the flux...

I guess if i was going with an inert gas i could probably get away with some kind of nickel-chromium alloy... That's what the heating elements are made of after all.

Any material scientists or metallurgists have any ideas?

blogfast25 - 30-4-2014 at 12:17

Nice crystals. Grow bigger ones and a fortune could be in the making, even though they wouldn't fetch the same as natural ones.

deltaH - 1-5-2014 at 23:22

Yeah on second thoughts rskennylmore, I doubt zirconium will stand your extreme temperatures in an oxidising environment and while they are far less costly than platinum crucibles, the gamble probably ain't worth it considering that they are, nevertheless, not cheap!

rskennymore - 4-5-2014 at 07:58

I think I might try plumbing nitrogen into the furnace with a zirconium crucible. I found a source of 45ml Zirconium crucibles for around 50 bucks each and ordered one. If it doesn't work then it'll just go in the element collection.

Also, I've been doing some reading... It seems adding a layer of iridium to refractory metals has been studied to some extent by NASA: https://archive.org/details/nasa_techdoc_19680019183

According to their results section (page 43 of the pdf) Niobium plated with Iridium showed no measurable amount of loss due to diffusion after 30 hours at 1300C. Since the experiment for aluminium oxide crystal growth requires 1100C, I think an Iridium plated Niobium crucible would probably last quite a while and still come in a lot cheaper than a platinum crucible.

They mention a lot of difficulties in achieving a coherent plating on the more reactive metals (niobium/tantalum). I don't think I have the resources to attempt the pressure-bonding method they employ, or the molten salt electrolyte. However this patent: US20120073980 gives some promise for a diy-er.

At one point, the NASA document mentions that Iridium electroplates easily onto nickel and that a intermediary layer of nickel enabled them to plate a coherent layer onto tantalum and niobium.

Zirconium isn't mentioned in any of these documents.

Much experimentation to come.

deltaH - 4-5-2014 at 09:41

Wow, 50 bucks... that doesn't sound too bad. I guess it is worth the gamble if you can get nitrogen in there. Best of luck with the trial!

As for iridium plating, I have a sneaky suspicion that it might just be one of those things that doesn't sound too complicated in principle... :D

rskennymore - 17-5-2014 at 15:15

So, after reading a bit about zirconium, I learned it forms a nitride so I opted for a 40 cubic foot tank of argon and pumped it into the furnace at a rate of about 5 cubic feet per hour.

Evidently this isn't enough as after running the experiment I found a familiar crumbled heap of zirconium oxides. I have yet to sift through it for ruby crystals but poking through the remains didn't turn up anything any bigger than what I saw with the first attempt using a zirconia crucible.

I'll post more information when it cools enough to sift through, but I'm starting to think this method of crystallizing alumina will require very specialized materials.




deltaH - 18-5-2014 at 00:20

That's a pity... sorry to hear you got no joy from the zirconium!

rskennymore - 18-5-2014 at 09:31

So... The Zirconium went Chernobyl on me, though there are those tiny grains of ruby sand in there again.

Due to the relatively high cost of argon compared with nitrogen, I think I will focus on materials that are inert under nitrogen. I am wondering about graphite now...

deltaH - 18-5-2014 at 10:32

I think the molybdenum trioxide is too strong an oxidant for graphite... it's almost certainly going to go chernobyl too (your fusing an oxidising agent with an excellent fuel at 1000°C)!

I have discovered an interesting thing that has bearing on your experiments, aluminium molybdate, Al2(MoO4)3 is a known compound and has a low melting point (704°C) according to Wikipedia.

This is most likely the compound that forms when you fuse aluminium oxide and molybdenum trioxide by this reaction:

Al2O3(s) +3MoO3(l) <=> Al2(MoO4)3(l)

The implications of this is that it means you can actually dissolve a hell of a lot more alumina in molybdenum trioxide, because your actually forming molten aluminium molybdate!!!

This means that you could/should maybe use a lot more alumina than what the authors used to form a 'saturated' solution, well not really a saturated solution but actually just molten aluminium molybdate.

That said, perhaps then if forming aluminium molybdate, you could fire this in an alumina crucible, as now it's 'saturated' and so shouldn't dissolve more alumina.

I think this might also help to form much larger crystals.

So my suggestion is this, use a mole ratio of MoO3 to gamma Al2O3 powders of 3 so that expressly target the formation of aluminium molybdate and do this in a alpha alumina crucible.

Gamma alumina, being highly porous, should react/dissolve rapidly and preferentially with the molybdenum trioxide powder, whereas the alpha alumina crucible should be much more kinetically inert because it is highly crystalline and has a low surface area for reaction.

[Edited on 18-5-2014 by deltaH]

PHILOU Zrealone - 22-5-2014 at 10:50

The process seems fine, but I think the cistallization time and the size of the batch matters with regards to the size of the rubies obtained.

Wizzard - 22-5-2014 at 13:41

Hypothetical process:

If we were to use a tall crucible, and fill the bottom with a strong mix of alumina and molyb trioxide
Add alumina with decreasing molyb trioxide to the top, to some end ratio by the filling point which is less than the ratio of MolybT:Alumina.
Then, heating strongly (>1000*C) from the bottom and keeping the (loosely capped) top of the crucible cooler than the base(<750*C)

I think as the molyb trioxide liquefies/reacts and slowly boils off, it works its way up and liquefies/reacts with the column of alumina, and larger crystals may be formed as the denser solute falls and before all of the molyb trioxide is lost to the air, or works it's way out of the mixture/solution.

I agree with deltaH - I think more alumina might help. But there's a chance, only a blob of alumina will be the end result.

PHILOU Zrealone - 23-5-2014 at 11:38

Quote: Originally posted by PHILOU Zrealone  
The process seems fine, but I think the cistallization time and the size of the batch matters with regards to the size of the rubies obtained.

I have long thought about it after seing a TV emission about 20 years ago where they have succeded in making big cristals of ruby; they where cristal clear with faces and about 2cm long on 1,5 large... almost like homemade candies.
The process they used was a secret melting media very close to the natural one. The price of those artificial gems was 1/20 of the price of the natural ones. Since the media was very close to the natural one, distinction between natural and artificial was very difficult, so to be fair, they added dopping stuffs that allowed orange fluorescence under UV light or discrete radiations.

I have tried the melting process with acetylen-oxygen blowtorch (goes upt to 3100°C in the flame cone)...but with such gas processes one gets bubble inclusions and carbon inclusions. I succeeded in doing tiny ruby spheres...but most of the Al2O3-Cr2O3 mix goes into the air while blowing.
To avoid this one could use a large Fresnel lens: this process is used to melt sand and allow to make 3D glass solar printer. With the sun beam, no gas, only heat.

For the rest red corundum (ruby) can be used as laser pumper, or as special mirror because of its very high refractive index.

Corundum can be all colors by varying the trace amounts of colourizer oxyd.
No --> leuco corundum or colorless
Fe2O3 --> Yellow
Cr2O3 --> Red
CoO --> Blue (saphire)
CuO --> Green
FeO --> Blue
TiO2 --> Blue
By mixing one can get all colors from the rainbow and by lowering the quantity of colorizer one get very light shades.

The melting process like Verneuil's make cylinders that breaks irregularly into the lenght like glass. Those are used for training in jewelry.

I have some natural rubies and artificial ones... il will try to put a picture.

[Edited on 23-5-2014 by PHILOU Zrealone]

deltaH - 23-5-2014 at 23:34

I would think that pink coloured gemstones would be sough after, even as synthetics. That would be a nice challenge for the hobby gemologist and one could presumably sell pink stones on ebay for a modest premium over other synthetics.

The question is how to dope for pinks?

Some possibilities: Mn2+, Nd3+, Er3+??
If it was me, I'd try an erbium(III) oxide for a pretty fluorescent pink as the others may give varied results depending on the specifics of the crystal habit.

[Edited on 24-5-2014 by deltaH]

rskennymore - 24-5-2014 at 05:29

As to adding more alumina, one of my attempts with a silicon carbide crucible I did try increasing the amount. I was left with basically blobs of fused alumina. This might be a useful method of making fused alumina ceramics at lower temperatures, but it isn't very useful for making rubies.

My theory was there were too many nucleation sites for it to grow crystals of any visible size. Also these runs were done after I had cooked my first thermocouple and was using the furnace's built-in thermocouple which isn't as accurate. It's possible the target temperature of 1100C was not reached for long enough to ensure complete dissolution. I still think there was too much Alumina to dissolve though.

As to the formation of Aluminium Molybdate... If it does form it doesn't stick around at those temperatures. There is no mention of residue or difference in mass of collected ruby crystals to that of the solute added in any of the publications I have found.

Wizzard's suggestion seems similar to the hydrothermal method, albiet a bit backwards. In the hydrothermal method, instead of Molybdenum Trioxide, an aqueous solution of alkali hydroxides is used as the solvent at increased pressure and temperature. From what I have read, around 800 degrees in a sealed steel "bomb". The seed crystals are placed at the top, which is kept cooler than the bottom, which is where the crystal "nutrients" are kept. This is the method used to grow quarts for electronics purposes, and some of the NdYAG laser crystals and various other industrial crystals. Wizzard, I suspect the cooler section would be where any crystals forming would grow, since the solubility of Alumina in the flux would be greatly diminished. Doesn't mean it wouldn't work in reverse though...

PHILOU Zrealone: The characteristic blue color of sapphire comes from the mixture of Iron oxide and titanium oxide, and not from CuO. As I understand it, it's the result of some complex physics having to do with the interaction of the two metal ions. I am not sure what color copper ions would impart, but once I figure out the whole crucible problem I would love to find out :)

deltaH: Couldn't you just limit the amount of chromium ions you add to get pink? Also looking at a few gemstone videos on youtube, I get the impression the pink ones aren't exactly rare or sought after. Even the "natural" ones. Apparently they heat-treat them to oblivion so they aren't much better than the Vernuli flame fusion rocks.

An update to the iridium crucible plan, I have obtained some Iridium sponge, and the chemicals needed to get it into solution. I haven't exactly got a plan as to what to plate it onto yet, however. I have a small lump of Niobium and Nickel plating equipment, so that might be a cheap test.

I've been researching and I think my first choice for an actual crucible will be Molybdenum. Since it's been demonstrated that plating of Iridium directly onto Molybdenum is possible without the need for the Nickel layer.

ww.mtialbany.com/wp-content/uploads/PDFs/A-72000-007-01-Molybdenum-Crucible-Price-List.pdf

45ml crucible for $415 USD...

At these prices though, not sure if it's worth it.

Molybdenum wire seems widely available on ebay, maybe I could plate onto that as a test and have a decent electrode for electrolysis for a perchlorate cell later on.

I'll start a thread on Iridium plating in the techno chemistry section when I get around to undertaking that mess.

Think I need to get into grad school... My basement lab is ill equipped for this level of metallurgy.

[Edited on 24-5-2014 by rskennymore]

PHILOU Zrealone - 24-5-2014 at 12:30

Quote: Originally posted by rskennymore  


PHILOU Zrealone: The characteristic blue color of sapphire comes from the mixture of Iron oxide and titanium oxide, and not from CuO. As I understand it, it's the result of some complex physics having to do with the interaction of the two metal ions. I am not sure what color copper ions would impart, but once I figure out the whole crucible problem I would love to find out :)

deltaH: Couldn't you just limit the amount of chromium ions you add to get pink? Also looking at a few gemstone videos on youtube, I get the impression the pink ones aren't exactly rare or sought after. Even the "natural" ones. Apparently they heat-treat them to oblivion so they aren't much better than the Vernuli flame fusion rocks.

[Edited on 24-5-2014 by rskennymore]

If you read exactly what I wrote:
"Corundum can be all colors by varying the trace amounts of colourizer oxyd.
No --> leuco corundum or colorless
Fe2O3 --> Yellow
Cr2O3 --> Red
CoO --> Blue (saphire)
CuO --> Green
FeO --> Blue
TiO2 --> Blue
By mixing one can get all colors from the rainbow and by lowering the quantity of colorizer one get very light shades. "

You will see that Cobalt oxyde (CoO - what is also used as blue glass colorizer) alone but also Iron oxyde (II) and Titanium dioxyde (alone or in mix) are responsible of the blue color of saphires and not Copper oxyde (CuO) what would make a green variant also obtainable by mixing TiO2 and/or FeO and/or CoO (for the blue) and Fe2O3 (for yellow) ...
-->Blue + Yellow = Green!

Since we have Red-Yellow-Blue all colours are reachable.

Fire rubies (red with a slight orange shade) can be obtained with traces of Cr2O3 and even lower trace amount of Fe2O3
-->Red+Yellow = Orange

As mentionned lowering the trace amount will give light shades... Red diluted = Pink. :cool:

Gemmology is one of my passions aside of chemistry...it is the expression of beauty of chemistry in the nature. I now have quite a big collection (I collect precious and less precious gemstones for about 30 years now).

[Edited on 24-5-2014 by PHILOU Zrealone]

Wizzard - 24-5-2014 at 16:54

When I get some time, I'll be giving my chemicals a mix and throwing them in the oven. Scary to think, my second favorite cup of crystals is *vastly* dwarfed in value by the container, pictured here - Pt, courtesy of my good friend Ivan (whom I owe some serious time).



Full Size

[Edited on 5-25-2014 by Wizzard]

[Edited on 5-25-2014 by Wizzard]

rskennymore - 25-5-2014 at 03:43

Quote: Originally posted by PHILOU Zrealone  

If you read exactly what I wrote:
"Corundum can be all colors by varying the trace amounts of colourizer oxyd.
No --> leuco corundum or colorless
Fe2O3 --> Yellow
Cr2O3 --> Red
CoO --> Blue (saphire)
CuO --> Green
FeO --> Blue
TiO2 --> Blue
By mixing one can get all colors from the rainbow and by lowering the quantity of colorizer one get very light shades. "

You will see that Cobalt oxyde (CoO - what is also used as blue glass colorizer) alone but also Iron oxyde (II) and Titanium dioxyde (alone or in mix) are responsible of the blue color of saphires and not Copper oxyde (CuO) what would make a green variant also obtainable by mixing TiO2 and/or FeO and/or CoO (for the blue) and Fe2O3 (for yellow) ...
-->Blue + Yellow = Green!

Since we have Red-Yellow-Blue all colours are reachable.

Fire rubies (red with a slight orange shade) can be obtained with traces of Cr2O3 and even lower trace amount of Fe2O3
-->Red+Yellow = Orange

As mentionned lowering the trace amount will give light shades... Red diluted = Pink. :cool:

Gemmology is one of my passions aside of chemistry...it is the expression of beauty of chemistry in the nature. I now have quite a big collection (I collect precious and less precious gemstones for about 30 years now).

[Edited on 24-5-2014 by PHILOU Zrealone]


You did say Cobalt, I apologize. I was trippin, as they say.

PHILOU Zrealone - 25-5-2014 at 13:08

Here are some artificial corundum I have aquired last year:
For reference of the size I have put a 1.5V AA Battery aside.
All where made via the Verneuil's process in the year 50-60.
-The big cylinder is about 3cm diameter and was normaly planned for optical/laser applications it is red-violet and costed like 60€.
-The orange and pink ones are half "bottles" or half "carots" costed each 3€.
-The tiny facetted rubies costed each 0,5€.

General overview in direct sunlight and under the flash of the camera:
Nat-1.JPG - 219kB Flash-1.JPG - 209kB

Closer view in direct sunlight and under the flash of the camera:
Nat-2.JPG - 171kB Flash-2.JPG - 210kB

[Edited on 25-5-2014 by PHILOU Zrealone]

deltaH - 25-5-2014 at 23:54

Stunning PHILOU, very nice photos!

TheAlchemistPirate - 26-5-2014 at 09:35

I am looking into making synthetic rubies too, but I don't know why we cant just use the Czochralski process with alumina and chromium oxide. I figured I could build a firebrick electric furnace capable of reaching 2075 degrees Celsius, put it a graphite crucible and seal the space around the top with firebrick and high-heat insulation foam. Then I would put a graphite rod in attached to a pulley slowly rotating it in the crucible, and let the crystals collect on the rod. Wouldn't this work? I know this sounds dumb and obviously if it was this simple we would all be making rubies, but please explain.

Morgan - 15-7-2014 at 07:14

iPhone 6 Sapphire Crystal Display!
https://www.youtube.com/watch?v=5R0_FJ4r73s#t=1m23s